Question-Answering Method, System, and Program for Answering Question Input by Speech

1. A question answering method for answering a question by using a text database storing text data in conjunction with a speech database storing speech data, comprising: receiving a speech signal of the question; performing a speech recognition for the speech signal to obtain a speech recognition result including a recognition accuracy evaluation value; selecting one of the text database and the speech database by comparing the recognition accuracy evaluation value with a threshold value; searching the selected database for a search result; and generating an answer to the question based on the search result; wherein the speech recognition result includes text data, and wherein: if the recognition accuracy evaluation value is not less than the threshold value, then the text database is selected and searched by using the text data as a query term; and if the recognition accuracy evaluation value is less than the threshold value, then the speech database is selected and searched based on information received from the speech signal.

2. The method according to claim 1 , further comprising recognizing a questioner from the speech signal, and wherein the answer is generated by using a part of the search result corresponding to the recognized questioner in preference to another part of the search result corresponding to a person other than the questioner.

3. The method according to claim 1 , further comprising recognizing a questioner from the speech signal, and wherein the answer is generated by using an authorized part of the search result to which the recognized questioner has access permission.

4. The method according to claim 1 , further comprising: displaying a plurality of text data or a plurality of speech data obtained as the search result from the text database or speech database; and deleting, from the text database or speech database, the same data as one of the plurality of displayed text data or speech data in accordance with a user's request.

5. The method according to claim 1 , further comprising: displaying a correction window which represents the speech recognition result; and correcting the text database in accordance with a user's correction instruction in the correction window.

6. The method according to claim 1 , further comprising: displaying a correction window which represents the speech recognition result; and correcting a speech recognition dictionary for use in speech recognition in accordance with a user's correction instruction in the correction window.

7. A system for answering to a question, comprising a computer coupled to a text database storing text data and a speech database storing speech data, the computer including question answering software which configures the computer to: receive a speech signal of the question; perform a speech recognition for the speech signal to obtain a speech recognition result including a recognition accuracy evaluation value; select one of the text database and the speech database by comparing the recognition accuracy evaluation value with a threshold value; search the selected database for a search result; and generate an answer to the question based on the search result; wherein the speech recognition result includes text data, and wherein the question answering software further configures the computer to: select and search the text database, using the text data as a query term, if the recognition accuracy evaluation value is not less than the threshold value; and select and search the speech database based on information received from the speech signal, if the recognition accuracy evaluation value is less than the threshold value.

8. The system according to claim 7 , wherein the question answering software further configures the computer to: recognize a questioner from the speech signal; and generate the answer by using a part of the search result corresponding to the recognized questioner in preference to another part of the search result corresponding to a person other than the questioner.

9. A system according to claim 7 , wherein the question answering software further configures the computer to: recognize a questioner from the speech signal; and generate the answer by using an authorized part of the search result to which the recognized questioner has access permission.

10. The system according to claim 7 , further comprising a display device coupled to the computer, and wherein the question answering software further configures the computer to: display, on the display device, a plurality of text data or a plurality of speech data obtained as the search result from the text database or speech database; and delete, from the text database or speech database, the same data as one of the plurality of displayed text data or speech data in accordance with a user's request.

11. The system according to claim 7 , further comprising a display device coupled to the computer, and wherein the question answering software further configures the computer to: display, on the display device, a correction window which represents the speech recognition result; and correct the text database in accordance with a user's correction instruction in the correction window.

12. The system according to claim 7 , further comprising a display device coupled to the computer, and wherein the question answering software further configures the computer to: display, on the display device, a correction window which represents the speech recognition result; and correct a speech recognition dictionary for use in speech recognition in accordance with a user's correction instruction in the correction window.

13. A computer-readable recording medium encoding a program executed by a computer for answering a question by using a text database storing text data in conjunction with a speech database storing speech data, the program comprising: means for instructing a computer to receive a speech signal of the question; means for instructing the computer to perform a speech recognition for the speech signal to obtain a speech recognition result including a recognition accuracy evaluation value; means for instructing the computer to select one of the text database and the speech database by comparing the recognition accuracy evaluation value with a threshold value; means for instructing the computer to search the selected database for a search result; and means for instructing the computer to generate an answer to the question based on the search result; wherein the speech recognition result includes text data, and wherein: if the recognition accuracy evaluation value is not less than the threshold value, then the text database is selected and searched by using the text data as a query term; and if the recognition accuracy evaluation value is less than the threshold value, then the speech database is selected and searched based on information received from the speech signal.

14. A computer-readable recording medium according to claim 13 , the program further comprising means for instructing the computer to recognize a questioner from the speech signal, and wherein the answer is generated by using a part of the search result corresponding to the recognized questioner in preference to another part of the search result corresponding to a person other than the questioner.

15. A computer-readable recording medium according to claim 13 , the program further comprising means for instructing the computer to recognize a questioner from the speech signal, and wherein the answer is generated by using an authorized part of the search result to which the recognized questioner has access permission.

16. A computer-readable recording medium according to claim 13 , the program further comprising: means for instructing the computer to display a plurality of text data or a plurality of speech data obtained as the search result from the text database or speech database on a display device; and means for instructing the computer to delete, from the text database or speech database, the same data as one of the plurality of displayed text data or speech data in accordance with a user's request.

17. A computer-readable recording medium according to claim 13 , the program further comprising: means for instructing the computer to display a correction window which represents the speech recognition result on a display device; and means for instructing the computer to correct the text database in accordance with a user's correction instruction in the correction window.

18. A computer-readable recording medium according to claim 13 , the program further comprising: means for instructing the computer to display a correction window which represents the speech recognition result on a display device; and means for instructing the computer to correct a speech recognition dictionary for use in speech recognition in accordance with a user's correction instruction in the correction window.